Predictive analytics for cost modeling challenges in polymer industry
Polymer manufacturing is a $100B+ industry that includes more than 7800 companies in the US. Plastics and polymers are a significant export market and the midwest region alone accounts for more than 100,000 jobs in this industry. Given the critical nature of this business, optimizing business processes is almost a necessity and is an area where advanced analytics can add significant value.
There are several areas within the polymer supply chain where analytics can add real value. Improving forecast accuracy, reducing inventory levels and optimizing distribution can provide working capital improvement, for example. The first step is to gain a high level overview of the supply chain and leverage analytics to find and improve the most critical areas for working capital improvements.
Fluctuations in feedstock costs arise due to a variety of reasons, including weather related effects. When raw material costs fluctuate, profitability can be adversely impacted, and unless the effects of raw material costs on the final product are properly modeled, it is difficult to optimize the supply chain. Analytics can help in accurately modeling the impact of these variations on the final product cost.
Analytics dashboards can provide a clear strategic view using up-to-date information to tell the story of how the company has performed over the last month, quarter or year, as well as using forward-looking predictive analytics to show executives where the supply chain is headed. Industry leading companies employ dashboards and KPIs to continually monitor operational performance. Other typical supply chain challenges exist in optimizing financial performance, improving sales and operations planning and sourcing.
Logistics and distribution in polymer industry poses a very unique challenge. The customer for a typical polymer manufacturer is another manufacturer, for example, a shampoo or soft drink bottle manufacturer, who in turn sells the processed product (e.g. bottle) to a consumer product company. The consumer is final end user of the finished product and is four levels removed from the polymer manufacturers. Transportation costs are rolled up into inventory costs for a polymer manufacturer and thus transportation cost modeling and forecasting becomes a significant challenge.
In all of these cases, there is a need to understand cause and effect, detect key drivers of business performance, develop accurate cost models which relate input factors to the finished product and build good forecasting ability. Using already available data, predictive analytics help address all of these areas which impact the business performance of a polymer manufacturer.
Download our case study to learn how one manufacturer is leveraging analytics to improve costs.